南岭国家级自然保护区森林景观格局变化与动态模拟

Forestry Landscape Patterns Changes and Dynamic Simulation of Nanling National Nature Reserve, Guangdong

基于eCogntion、ArcGIS和IDRISI软件,采用景观格局指数分析广东南岭国家级自然保护区1988~2009年景观类型数量及空间格局的变化;运用CA-Markov模型模拟流域2010年的景观格局,预测2021年的景观格局。结果表明,研究区森林景观类型以常绿阔叶林和针叶林为主;景观破碎度增加,斑块复杂程度提高,各景观类型的分布更加趋于复杂化;CA-Markov模型预测表明,2010~2021年景观破碎度有所降低,多样性增加。

Forest landscape pattern and its change is a comprehensive reflection of interaction with natural and artificial factors. In this article, landscape pattern changes of Nanling National Nature Reserve from 1988 to 2009 were analyzed with landscape pattern index based on eCognition and ArcGIS. CA-Markov model was ad- opted to simulate the landscape pattern in 2010 and to predict the landscape pattern in 2021. The results showed that the study area was mainly forest-based landscapes, more than 95% of the total area, in which, ever- green broad-leaved forest and coniferous forest were two dominant landscape types, other landscape types were distributed in the study area with scattered or gathered, and forming a rich landscape mosaic pattern. From 1988 to 2009, the area of coniferous forest decreased whereas areas of other land and construction land increased. Coniferous mixed forest, evergreen broad-leaved forest, broad-leaved mixed forest and farmland first expanded and then shrank, whereas shrubs, other forests and waters first shrank and then expanded. The level of landscape fragmentation and plaque complexity was increased, the proportion of landscape type has a larger change; landscape diversity was increased after reduced, the dominance, evenness changed little, remain- ing stable, the fractal dimension of landscape patches was different at different periods, and the increase ten- dence of human disturbance increased complex to the conversion of various types of plaque. Analysis land- scape types changes of time and space, which found that the largest area ratio changes between with coniferous forest and other woodlands, evergreen broad-leaved forest and broad-leaved mixed forest transformation was frequent, also conversion frequency between mixed coniferous forest and coniferous forest, shrub and other landscape types. Construction land was mainly transformed from evergreen broad-leaved forest and coniferous forest, water, as a landscape type in the study area, mainly transformed by coniferous forest. From 2010 to 2021, with CA-Markov model, it was predicted that the area of coniferous forest and evergreen broadleaf for- est would reduce, and other landscape trend would grow, of which construction land and water would grow more rapidly. Also the number of major forest landscape patches would reduce significantly, and the distribu- tion be more concentrated, which means the degree of aggregation and connection would increase, fragmenta- tion reduce, and landscape types distribution be towards uniformly in the future. And the level of landscape fragmentation would decline, and the level of landscape diversity would increase in the study area.